Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Fuzzy programming techniques for Computer Chess?

Author: Robert Hyatt

Date: 18:45:41 01/31/00

Go up one level in this thread


On January 31, 2000 at 09:19:50, Vincent Diepeveen wrote:

>On January 31, 2000 at 09:02:29, Robert Hyatt wrote:
>
>>On January 31, 2000 at 05:06:38, Harald Faber wrote:
>>
>>>
>>>Does any programmer use fuzzy programming within his program?
>>>If not, would it be helpful and make it easier or better to evaluate positions?
>>>AFAIK fuzzy is an ideal tool for combining several different, even complementary
>>>evaluation parameters, and chess programming has a large number of evaluation
>>>functions...
>>>
>>>Opinions?
>>
>>
>>I would suspect that _all_ chess programs have some 'fuzzy logic' in them.
>>one example is recognizing the stonewall pattern.  If you require the exact
>>pawn structure, you get zapped if he forgets to push the c2 pawn to c3 for
>>example.  If you 'fuzzy match' one pawn missing doesn't invalidate the
>>pattern.
>
>I see clear difference between using some form of fuzzy logic, cq see
>current chessprograms as a form of fuzzy logic, and making usage
>of 100% fuzzy programming.
>
>The methods as decribed in fuzzy logic programming are to say with an
>understatement 'naive', or even better 'quickly done to produce a paper'.
>
>Vincent
>
>>I certainly do it...


Nobody said 100% fuzzy logic programming.  I don't do that.  I do use fuzzy
logic in a few places when I want to "almost match" something, for example.



This page took 0 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.